CLONING AND SEQUENCING OF THE GENE ENCODING GLUTAMINE SYNTHETASE-I FROM THE ARCHAEUM PYROCOCCUS-WOESEI - ANOMALOUS PHYLOGENIES INFERRED FROM ANALYSIS OF ARCHAEAL AND BACTERIAL GLUTAMINE SYNTHETASE-I SEQUENCES

The gene glnA encoding glutamine synthetase I (GSI) from the archaeum Pyrococcus woesei was cloned and sequenced with the Sulfolobus solfata ricus glnA gene as the probe. An operon reading frame of 448 amino aci ds was identified within a DNA segment of 1,528 bp. The encoded protei n was 49% identical with the GSI of Methanococcus voltae and exhibited conserved regions characteristic of the GSI family. The P. woesei GSI was aligned with available homologs from other archaea (S. solfataric us, M. voltae) and with representative sequences from cyanobacteria, p roteobacteria, and gram-positive bacteria. Phylogenetic trees were con structed from both the amino acid and the nucleotide sequence alignmen ts. In accordance with the sequence similarities, archaeal and bacteri al sequences did not segregate on a phylogeny. On the basis of sequenc e signatures, the GSI trees could be subdivided into two ensembles. On e encompassed the GSI of cyanobacteria and proteobacteria, but also th at of the high-G+C gram-positive bacterium Streptomyces coelicolor (al l of which are regulated by the reversible adenylylation of the enzyme subunits); the other embraced the GSI of the three archaea as well as that of the low-G+C gram-positive bacteria (Clostridium acetobutilycu m, Bacillus subtilis) and Thermotoga maritima (none of which are regul ated by subunit adenylylation). The GSIs of the Thermotoga and the Bac illus-Clostridium lineages shared a direct common ancestor with that o f P. woesei and the methanogens and were unrelated to their homologs f rom cyanobacteria, proteobacteria, and S. coelicolor. The possibility is presented that the GSI gene arose among the archaea and was then la terally transferred from some early methanogen to a Thermotoga-like or ganism. However, the relationship of the cyanobacterial-proteobacteria l GSIs to the Thermotoga GSI and the GSI of low-G+C gram-positive bact eria remains unexplained.